A Predictive Model for Plant Response to Interactive Effect of Salinity and Phosphorous

Soil salinity and nutrient imbalances are two important limiting factors for plant production in arid and semiarid regions. Quantification of such limiting factors is pivotal for any management in these areas. In this investigation novel deterministic linear and nonlinear models were derived for predicting plant response to interactive effects of salinity and phosphorus through integration of a non-linear salinity model with two basic soil fertility models. To examine and evaluate performances of newly proposed models, several experiments were conducted under separate and combined salinity and phosphorus levels in field conditions. Treatments were consisted of five natural salinity (0.3, 3, 6, 9 and 12 dS m−1) and four phosphorus levels (0, 20, 40 and 80 mg kg−1 soil as KH2PO4), each with three replicates. The obtained results indicated that the proposed nonlinear model (Eq. 7) with R2 of 0.92 provides better estimations than other proposed models for canola yield under different levels of salinity, different levels of soil phosphorus and under combination of different levels of salinity and phosphorus. All proposed models accounting for simultaneous salinity and phosphorus showed that applying 20 and 40 mg P kg−1 soil, or even no use of phosphorus in soil had no significant impact on performance of salinity threshold value, while adding 80 mg P kg−1 soil would lead to increase the salinity threshold value.